Picture for Kewei Tu

Kewei Tu

Detecting Knowledge Boundary of Vision Large Language Models by Sampling-Based Inference

Add code
Feb 25, 2025
Viaarxiv icon

A Systematic Study of Cross-Layer KV Sharing for Efficient LLM Inference

Add code
Oct 18, 2024
Figure 1 for A Systematic Study of Cross-Layer KV Sharing for Efficient LLM Inference
Figure 2 for A Systematic Study of Cross-Layer KV Sharing for Efficient LLM Inference
Figure 3 for A Systematic Study of Cross-Layer KV Sharing for Efficient LLM Inference
Figure 4 for A Systematic Study of Cross-Layer KV Sharing for Efficient LLM Inference
Viaarxiv icon

Efficient Long-range Language Modeling with Self-supervised Causal Retrieval

Add code
Oct 02, 2024
Viaarxiv icon

Learning Robust Named Entity Recognizers From Noisy Data With Retrieval Augmentation

Add code
Jul 26, 2024
Figure 1 for Learning Robust Named Entity Recognizers From Noisy Data With Retrieval Augmentation
Figure 2 for Learning Robust Named Entity Recognizers From Noisy Data With Retrieval Augmentation
Figure 3 for Learning Robust Named Entity Recognizers From Noisy Data With Retrieval Augmentation
Figure 4 for Learning Robust Named Entity Recognizers From Noisy Data With Retrieval Augmentation
Viaarxiv icon

Dependency Transformer Grammars: Integrating Dependency Structures into Transformer Language Models

Add code
Jul 24, 2024
Viaarxiv icon

Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers

Add code
Jun 24, 2024
Figure 1 for Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers
Figure 2 for Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers
Figure 3 for Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers
Figure 4 for Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers
Viaarxiv icon

Unsupervised Morphological Tree Tokenizer

Add code
Jun 21, 2024
Figure 1 for Unsupervised Morphological Tree Tokenizer
Figure 2 for Unsupervised Morphological Tree Tokenizer
Figure 3 for Unsupervised Morphological Tree Tokenizer
Figure 4 for Unsupervised Morphological Tree Tokenizer
Viaarxiv icon

Layer-Condensed KV Cache for Efficient Inference of Large Language Models

Add code
May 17, 2024
Viaarxiv icon

Potential and Limitations of LLMs in Capturing Structured Semantics: A Case Study on SRL

Add code
May 10, 2024
Viaarxiv icon

RoT: Enhancing Large Language Models with Reflection on Search Trees

Add code
Apr 11, 2024
Viaarxiv icon